This subproject is one of many research subprojects utilizing the resources provided by a Center grant funded by NIH/NCRR. The subproject and investigator (PI) may have received primary funding from another NIH source, and thus could be represented in other CRISP entries. The institution listed is for the Center, which is not necessarily the institution for the investigator. We are currently studying in the DinB homolog (Dbh) lesion bypass DNA polymerase from S. solfataricus and will extend our work to the human DinB homolog (Pol kappa) within the next year. In the next two years we will focus on two aspects of these lesion-bypass polymerases:a. Error specificity on undamaged DNA templates. The Dbh polymerase makes -1 frameshift deletions at a high frequency when it encounters specific hotspot sequences. These hotspots consist of a run of 2 or 3 pyrimidines in the template strand flanked on the 5 side by a guanosine. When the enzyme encounters these hotspots, it skips one of the templating pyrimidines 50% of the time. This tremendously high error rate is probably linked to the enzymes preference for incorporating deoxycytidine (10-fold preference compared to the other 3 nucleotides). We want to understand the structural mechanism by which these errors are generated and the sequence specificity for skipping template pyrimidines and for preferentially incorporating cytidine into the primer strand.We have crystallized and solved (by molecular replacement) the structures of Dbh complexed with DNA containing a skipped template base at two different positions. In one position, just opposite the active site, the base is bulged out of the primer-template duplex where it is only in contact with solvent, not protein. In the other position, representing a stage where two nucleotides have been incorporated after the base was skipped, the base is bulged out of the primer-template duplex where it is bound in a small pocket on the enzyme (see attached figure). We are currently refining these structures to high resolution limits of 3.0 and 2.7 .Our goals over the next year are to determine co-crystal structures where the bulged base has been changed to each of the other 3 nucleotides (our current structures have cytidine as the bulged base). These structures will provide information about the preference for the enzyme skipping pyrimidine template bases. By placing the bulged base at several different positions in the template strand, we hope to understand how -1 frameshift mistakes are first generated, and then propagated. Additionally, we plan to determine structures with each of the 4 incoming nucleotides pairing with a complementary templating base.b. Bypass of DNA template lesions. Dbh has recently been shown to bypass an N2-furfuryl adduct of deoxyguanosine efficiently by incorporating deoxycytidine opposite the damaged template base. In the second year of this work, we plan to investigate the specificity of the bypass of this lesion by determining co-crystal structures of Dbh complexed with DNA containing deoxyguanosine template bases with various N2-adducts.